A surprising report on the funding of postgraduate degree programmes may have underestimated the true price of providing world-beating research and teaching in England's universities. Amy McLellan reports

A recent report highlighting the costs of postgraduate research degree programmes has sent a little-reported shiver through the higher education community. The report, produced by Bristol-based JM Consulting for the Higher Education Funding Council in England (Hefce), includes a warning that its findings "will probably be an unwelcome surprise to many in the sector".

A recent report highlighting the costs of postgraduate research degree programmes has sent a little-reported shiver through the higher education community. The report, produced by Bristol-based JM Consulting for the Higher Education Funding Council in England (Hefce), includes a warning that its findings "will probably be an unwelcome surprise to many in the sector".

Anecdotally, it does seem the headline figures have come as a shock to some institutions. Indeed, those of a nervous disposition should probably look away now. According to the report, the cost each year of a full-time postgraduate research student is £29,106 for those studying high-cost laboratory subjects; £23,815 for those on part-lab programmes; and £17,461 for library-based subjects. Income per student, almost without exception, falls well short of these totals.

JM Consulting reached these numbers by studying the time supervisors, examiners and lecturers spent on their postgraduate research students, taking into account not only the salary costs but also the associated indirect and estates costs. Consumables, such as laboratory chemicals or specialist software, and the costs of central services, student support, scholarships, bursaries and fees remission were also priced in.

Not everyone is surprised. "Even a back of the envelope calculation quickly adds up", says Howard Green of the UK Council of Graduate Education, and senior adviser to the vice-chancellor of Staffordshire University. "If you just look at the examination of a PhD student, that involves two to three days reading a thesis, a day on a viva and another day and a half writing a report, adding up to four and half days at around £30-£40 an hour. And those are just the direct staff costs with no account of the indirect costs faced by the university."

Some in the sector worry the actual costs could be even higher. The Hefce/JM Consulting figures focus on traditional full-time students and does not take into account part-timers, distance learners and a growing number studying for professional doctorates.

Janet Metcalfe of UK GRAD adds: "I'm not sure the study really recognises the costs of initiatives such as professional skills training or the implementation of the new QAA [Quality Assurance Agency] code of practice. It could be they've actually underestimated the true cost of running PhD research programmes."

The report admits its numbers are not hard and fast. A further study is recommended before the end of 2006, when improvements in institutions' costing systems should lead to more robust calculations.

Whether those calculations revise the numbers up or down, one thing is clear: there is a significant mismatch between funding and costs across the sector. Less clear is what the impact of exposing this mismatch will be on a sector already concerned about the concentration of research in a small band of élite institutions.

"In the long run, institutions and funders will have to take action to improve these levels of cost recovery if postgraduate research students are to become a financially sustainable part of UK higher education," said the report.

It does stress, however, that postgraduate research students figure in both the credit and debit columns of an institution's books. On the financial front, there are revenues directly linked to postgraduate research student registrations, including payments from Hefce, the Science Research Infrastructure Fund and fees. There are also indirect financial benefits: postgraduate research students can contribute to research that improves a department's Research Assessment Exercise rating, thereby boosting funding, or can free up others to undertake revenue-generating activities.

Postgraduate research students should also be given real credit for their non-financial contribution to the work, finances and credibility of an institution. They are a cheap source of labour for teaching and research, and represent the next generation of researchers and academics. A pool of talented research students contributes to an institution's prestige, attracts top-flight academics and generates a stimulating research environment.

Ralph Manley, director of graduate studies at Kingston University, says trying to put a price tag on the contribution of a postgraduate research student is like trying work out the cost of a tree. "Its timber may be worth $1,000 but when you try to work out its ecological value you cannot quantify it, and it's a little bit like that with research students," he says.

The Hefce/JM Consulting numbers are likely to focus minds on some of these non-financial benefits. They are also likely to prompt some soul-searching as institutions consider why they run research degree programmes, how effective those programmes are and how they fit with their overall strategy. This could stimulate a fundamental review of doctoral provision in this country.

"No one has a view about how many doctoral students we need," says Green, "and how much money we should put into the system. Other countries such as France and Canada do attempt this analysis but we don't have it here. This research might throw that discussion wide open again."

For now, the jury is still out. "More transparency has to be a good thing and it's really useful that Hefce has started work on this," says Metcalfe. "But there's still a lot more work to be done before we understand the full costs or the full implications of this."

Tom Sastry of the Higher Education Policy Institute (HEPI) has published a briefing paper as a follow-up to the Hefce/JM Consulting research. He believes it is impossible to predict how institutions will react now the costs of training and supervising postgraduate research students have been laid bare.

"It doesn't require any immediate action but four to five years down the line if an institution is struggling to balance its books then it might take a little bit of a different approach on research training," says Sastry.

One outcome could be a reduction in the number of research students as the supply of places shrinks and fee levels increase.

"We could see institutions charging higher fees and those students paying their own way could find the environment becomes less comfortable," says Sastry, adding that fee waivers - enjoyed by 15 per cent of home and EU students in 2003 - could become scarce in a more cost-conscious culture. Overseas students who pay high fees and work on projects a university expects to attract generous revenues are likely to find their places secured, although Sastry's report makes clear that even overseas students do not cover their costs.

Greater scrutiny of the non-financial benefits could change the profile of the typical postgraduate research student, with preference given to those who make a practical contribution to the work of the institution and whose research is closely integrated with that of the wider research team. As a result, those seeking to pursue research that does not fit a department's priorities or who want to study on a part-time or distance learning basis could find themselves squeezed out.

And, as HEPI's report makes clear, it is those institutions that do not attract Hefce research funding that face the biggest shortfall between revenue per student and cost per student. This funding gap could lead to a withdrawal of some institutions from postgraduate research student provision and accelerate the concentration of research training in those institutions where research quality is highest, a prospect that many in the sector fear would be detrimental to the quality, quantity and diversity of research in the UK.

Fact box

Theheadline cost to institutions per full-time student on a three-year programme was:

* Clinical and lab-based subjects (eg medicine, civil engineering): £87,317

* Part-lab subjects (eg maths, geography): £71,446

* Library-based subjects (eg business, classics): £52,383

The JM Consulting research broke down the costs of training and supervising postgraduates as follows:

* Time spent by supervisors, examiners and lecturers (salary costs): 13 per cent

* Time spent by supervisors, examiners and lecturers (indirect costs): 6 per cent

* Consumables, such as lab chemicals and specialist software: 31 per cent

* Scholarships, bursaries, fees remission: 9 per cent

* Indirect and estates costs, including central services, support time and costs, and the cost of capital employed adjustment: 40 per cent